Use of Preferred Source of Contraception Among Users of the Pill, Patch, and Ring in the US.

This survey study was reviewed and approved by the Committee for the Protection of Human Subjects at the University of California, Berkeley and the NORC Institutional Review Board. Survey methods were reported in accordance with American Association for Public Opinion Research ( AAPOR ) reporting guideline. From January to March 2022, we fielded a cross-sectional nationally representative survey through NORC’s AmeriSpeak Panel as part of a project to develop person-centered contraceptive access metrics. We developed and modified survey questions about contraceptive access based on an extensive literature review and interviews conducted with 36 stakeholders from various sectors (eg, reproductive justice, clinical care, advocacy, research). To evaluate the comprehensibility and accessibility of key survey questions, we conducted 33 cognitive interviews with individuals meeting survey eligibility criteria. We convened the Person-Centered Contraceptive Access Metric Working Group reflecting similar diversity to stakeholder interviews to develop the metrics, which have been detailed elsewhere. 37 Informed by the person-centered health care access 32 and sexual and reproductive health (SRH) equity frameworks, 38 the contraceptive access metrics address major gaps by centering self-identified need for contraceptive methods, services, and information. The AmeriSpeak panel primary sampling frame is the 2010 NORC National Frame, a multistage probability sample constructed to represent the US household population. AmeriSpeak includes 54 001 members from more than 43 000 households, covering an estimated 97% of US households. 39 NORC provided deidentified survey data. Eligibility criteria included being aged 15 to 44 years, assigned female sex at birth, not known to be sterile, and able to complete the survey in English or Spanish. NORC invited all panelists aged 18 to 44 years meeting sex criteria to read an online informed consent form and to consent by selecting yes. Those who consented completed the eligibility questionnaire and the full survey if eligible. For individuals aged 15 to 17 years, NORC first contacted parents; if they consented, NORC invited their child to assent and complete eligibility questions and the survey. Participants accessed the self-administered survey online with a median length of 25 minutes. Among eligible respondents who completed screening, the response rate was approximately 97%. NORC completed data quality checks, including removing respondents who completed the survey very quickly, refused to answer more than 50% of questions, and provided the same response across grid questions. Respondents received the equivalent of $8 in NORC points (NORC’s standard remuneration system for panelists) upon survey completion. Primary measures of interest are preferences for contraception sources and use of preferred sources. To assess preferences, we asked respondents, “If you could choose any way of getting a birth control prescription, how would you prefer to get it?” Response options were in-person at a doctor’s office or clinic; in-person at a hospital; from a telehealth (video or phone) appointment with a doctor’s office, clinic, or hospital; prescribed by a pharmacist; over the counter at a pharmacy or other store; from an online service that sends it directly; and in another way. Respondents could select multiple sources; those who did were prompted to select their most preferred source. For analysis, we combined the 2 in-person sources into a single category. We also asked respondents, “The last time you got [method], how did you get it prescribed?” Responses were the same as above, excluding the OTC option (as there was not yet an OTC pill), and adding a response capturing asynchronous telehealth care: “contacted my clinician and they wrote a new prescription without seeing me for any kind of appointment first.” To determine whether respondents had used a preferred source of contraception, we combined responses from the first question capturing their preferences and the second capturing where they had most recently obtained contraception to construct metrics that indicated receipt of contraception from any or the most preferred source. For example, if a respondent indicated multiple preferred sources (eg, in-person, pharmacist prescribing, and OTC) and had most recently obtained their method in-person, they were considered to have used any preferred source of contraception. However, if the same respondent most preferred receiving contraception OTC, they were not classified as using their most preferred source of contraception. If respondents indicated telehealth was a preferred source of contraception, they were classified as using a preferred source if they had mostly used synchronous (video or phone appointment) or asynchronous (contacting a clinician for a new prescription without an appointment) care (eAppendix in Supplement 1 ). We asked respondents why they preferred their top source of contraception (16 response options consolidated into 10 categories). For bivariate and regression analyses, we created a binary variable that indicated whether respondents most preferred traditional in-person care vs an alternative source (ie, telehealth, pharmacist-prescribed, OTC, online service). The survey also included sociodemographic items and SRH characteristics. NORC collected self-reported race and ethnicity as part of their standard demographic items for all AmeriSpeak panelists and provided it to the research team. Race and ethnicity categories included Asian or Pacific Islander only, Black only, Latina or Hispanic, White only, multiracial (not including Latina or Hispanic), or other race and ethnicity only. We do not have the specific race and ethnicity for the 7 respondents in this analytic sample who are categorized as other. We asked about experiences obtaining contraception, including missing a pill, patch, or ring because they were unable to get it on time; encountering challenges or delays in the past year; and whether they would have liked to change something about the process. Among respondents who had ever discussed contraception with a health care clinician, we included the person-centered contraceptive counseling measure. 40 Respondents who received person-centered counseling rated their most recent contraceptive care clinician as excellent on: respecting them as a person, letting them say what mattered to them about their birth control, taking their preferences about your birth control seriously, and giving them enough information to make the best decision about their birth control method. We also asked about discriminatory experiences due to race, ancestry, or national origin. 41 Those who had ever discussed contraception with a health care clinician were asked how often they had experienced the following types of discrimination in family planning settings during their lifetimes: being treated with less courtesy than other people; being treated with less respect than other people; receiving poorer service than other people; feeling like a doctor or nurse was not listening; a doctor or nurse assumed they were on welfare; a doctor or nurse made assumptions about the number of children they had; a doctor or nurse assumed they had multiple sexual partners; a doctor or nurse strongly encouraged them to use one method of birth control when they preferred another; and a doctor or nurse assumed they had a sexually transmitted disease, such as chlamydia, gonorrhea, genital warts, herpes, or HIV. We created a categorical variable describing the number of discrimination types ever experienced (ie, 0, 1 to 4, or 5 to 9). The analytic sample included individuals who currently used the contraceptive pill, patch, or ring and were not pregnant. We excluded 4 respondents reporting use of more than 1 of these methods in the past month, 11 respondents missing data on preferred source of contraception, and 7 respondents who most preferred another source that could not be categorized. The sociodemographic characteristics of the full sample and a bivariate analysis of those included vs excluded from the analytic sample are described in the eTable in Supplement 1 . We present descriptive statistics for sociodemographic and SRH characteristics, use and type of preferred contraceptive source, and reasons for preferences as unweighted numbers with weighted prevalences. We used Rao-Scott corrected χ 2 tests to examine the association between whether respondents preferred a traditional in-person or alternative source, reasons for their preference, and previous experiences with contraceptive care. Additionally, we used multivariable logistic regression to examine associations between previous experiences and a preference for accessing care through an alternative contraceptive source vs traditional in-person care. Regressions adjusted for age, race and ethnicity, sexual orientation, education, insurance type, employment status, urbanicity, parental status, current pill use, reasons for SARC use, and duration of SARC use. Due to cell size issues, we were not able to conduct analyses considering the specific types of alternative sources. In all analyses, we used survey weights (constructed by NORC) to account for differences between the survey sample of 3059 individuals and the US population based on age, education, race and ethnicity, marital status, number of children in household, and age by race and ethnicity. We conducted all analyses using Stata version 17.0 (StataCorp). We applied svy commands to account for weighting and complex survey design. All reported proportions are weighted. Statistical significance was set at P  < .05, and all tests were 2-sided. Data were analyzed from January 25, 2023, to August 15, 2024.

Among the full sample of 3059 individuals (unweighted), 2338 unweighted respondents (70.3%) had ever used the pill, patch, or ring (eTable in Supplement 1 ). Of the full sample, 595 unweighted respondents (21.7%) currently used these methods, comprising the analytic sample (data not shown). A lower proportion of parents and a higher proportion with commercial insurance were included in the analytic sample compared with those excluded (eTable in Supplement 1 ). Within the analytic sample, 126 unweighted individuals (weighted prevalence, 45.7%) were younger than 25 years ( Table 1 ). Of the 595 individuals included in the sample, most identified as female (581 unweighted individuals; weighted prevalence, 95.9%), straight or heterosexual (495 unweighted individuals; weighted prevalence, 77.8%), and had commercial health insurance (455 unweighted individuals; weighted prevalence, 72.1%). Additionally, the study included 72 unweighted Black individuals (weighted prevalence, 13.1%), 105 unweighted Latina or Hispanic individuals (weighted prevalence, 20.7%), and 350 unweighted White individuals (weighted prevalence, 58.4%). Abbreviations: CHIP, Children’s Health Insurance Program; IHS, Indian Health Service; VA, Veterans Affairs. We do not have the specific race and ethnicity for the 7 respondents in this analytic sample who are categorized as other.. Most respondents were pill users (543 unweighted respondents; weighted prevalence, 92.3%), and 360 unweighted respondents (weighted prevalence, 62.1%) indicated they were currently using their preferred contraceptive method. Half (296 unweighted respondents; weighted prevalence, 49.7%) had most recently obtained their SARC method through any preferred source, and 227 unweighted respondents (weighted prevalence, 39.8%) had obtained it through their most preferred source ( Table 2 ). Most commonly, respondents had last obtained their method in-person at a doctor’s office, clinic, or hospital (448 unweighted respondents; weighted prevalence, 73.8%), followed by an online service (48 unweighted respondents; weighted prevalence, 9.0%), asynchronous telehealth (44 unweighted respondents; weighted prevalence, 7.9%), synchronous telehealth (33 unweighted respondents; weighted prevalence, 5.2%), and prescribed by a pharmacist (13 unweighted respondents; weighted prevalence, 2.2%). Abbreviations: PCOS, polycystic ovary syndrome; SARC, short-acting reversible contraception; STI, sexually transmitted infection. The term birth control was used in the survey measure itself and is included in the tables and manuscript text when referring specifically to the survey language. In other instances, the term contraception is used. Responses not mutually exclusive. Excludes segesterone acetate/ethinylestradiol 1-year ring user. Respondents reported using their SARC method for pregnancy prevention (507 unweighted respondents; weighted prevalence, 84.2%), menstrual regulation (399 unweighted respondents; weighted prevalence, 71.6%), and to alleviate period pain (254 unweighted respondents; weighted prevalence, 46.2%); only 122 unweighted respondents (weighted prevalence, 15.7%) were exclusively using their SARC method for pregnancy prevention. Additionally, 448 unweighted respondents (weighted prevalence, 73.8%) had used their method for more than a year; 221 unweighted respondents (weighted prevalence, 32.2%) had used it for more than 5 years. A majority (498 unweighted respondents; weighted prevalence, 80.9%) had most recently received a 1-month to 3-month supply of their method. One hundred ninety-three unweighted respondents (weighted prevalence, 35.4%) had ever missed using their method because they were unable to get it on time; 128 unweighted respondents (weighted prevalence, 25.1%) had encountered challenges or delays obtaining contraception in the past year; and 90 unweighted respondents (weighted prevalence, 14.5%) wanted to change something about the process to obtain contraception. Just over a third (261 unweighted respondents; weighted prevalence, 38.1%) of respondents received person-centered contraceptive counseling from the most recent contraceptive care clinician. Half (289 unweighted respondents; weighted prevalence, 51.6%) had ever experienced discrimination in a family planning setting. Less than half (257 unweighted respondents; weighted prevalence, 44.7%) of respondents selected in-person, clinic-based care as a preferred source—followed by OTC (208 unweighted respondents; weighted prevalence, 32.4%), online services (165 unweighted respondents; weighted prevalence, 27.4%), pharmacist-prescribed (158 unweighted respondents; weighted prevalence, 26.4%), and telehealth (165 unweighted respondents; weighted prevalence, 25.4%) ( Table 3 ). When considering their most preferred source of contraception, about one-third (197 unweighted respondents; weighted prevalence, 35.6%) of respondents selected traditional in-person, clinic-based care. Reasons for selecting their most preferred source included convenience (380 unweighted respondents; weighted prevalence, 63.1%), saving time (286 unweighted respondents; weighted prevalence, 48.0%), higher quality of care (trusting the clinician, feeling respected; 181 unweighted respondents; weighted prevalence, 33.0%), and affordability (lower cost, covered by insurance; 174 unweighted respondents; weighted prevalence, 32.5%) ( Table 4 ). In bivariate analyses, those who selected convenience and saving time as reasons for their preference were more likely to most prefer an alternative source, whereas those who selected trusting and feeling respected by their clinician, usual location, and being able to ask questions were more likely to prefer traditional, in-person care. Responses not mutually exclusive. Respondents could select multiple reasons for their preference. In both bivariate and adjusted regression analyses, those who previously experienced issues related to obtaining contraception were more likely to prefer an alternative contraceptive source, while those who previously received higher quality care had reduced odds of preferring an alternative source ( Table 5 ). Specifically, there were reduced odds of greatest preference for an alternative contraceptive source among those who had recently received person-centered contraceptive counseling (adjusted odds ratio [aOR], 0.59; 95% CI, 0.35-0.98) and those who had ever experienced 1 to 4 types of discrimination in a family planning setting (aOR, 0.45; 95% CI, 0.21-0.95) compared with 5 to 9 types. Abbreviations: aOR, adjusted odds ratio; NA, not applicable; SARC, short-acting reversible contraception. Regressions adjust for age, race and ethnicity, sexual orientation, education, insurance type, employment status, urbanicity, status as a parent, current pill user, reasons for SARC use, and duration of SARC use.

About half of respondents who were currently using the contraceptive pill, patch, and ring were not obtaining their methods through a preferred source of contraception. Notably, only 35.6% of respondents most preferred to receive care in-person, the dominant model of contraceptive care delivery in the US. 11 , 12 , 13 Building upon previous research, we found that those who want a source that is more convenient and/or takes less time most prefer alternative sources. Those who previously experienced challenges accessing contraception were more likely to prefer alternative sources. Those who previously had received higher quality family planning care were less likely to prefer alternative sources, aligning with research documenting that nonperson-centered and discriminatory care alienates patients from the health care system. 37 , 42 , 43 With 80% of SARC users in the sample most recently receiving only a 1- to 3-month supply, the norm of using these methods requires frequent engagements with the health care system. The short length of supplies dispensed presents many barriers for continued use and suggests failure to implement policies in 23 states that require Medicaid and/or commercial insurers to cover an extended supply of contraception (usually 12 months). 44 It also indicates the presence of barriers constraining implementation of clinical guidelines that recommend routine dispensing of a 12-month supply of contraception. 44 , 45 Clinician-level barriers include lack of awareness of the practice, belief that a 12-month supply would not be reimbursed by insurance, and implicit bias influencing decisions about which patients can and cannot handle a 12-month supply. 46 A documented system-level factor is the need to update electronic health record templates so that a 12-month SARC supply is the default order length. 47 To our knowledge, this is the first national study assessing the use of preferred source of contraception using a comprehensive range of sources, including OTC. We use population-based, nationally representative data, so findings have a high level of generalizability. We developed the metric use of preferred source of contraception using a process grounded in values of person-centered health care access 32 and SRH equity. 38 Policy strategies to expand contraceptive access must address implementation challenges that stand in the way of expanded access—for instance, those documented for implementing pharmacist prescribing and the requirement to have a prescription for OTC contraception to be covered by insurance. 48 , 49 There is also a need for additional methods to be available OTC, assuming systems are in place for patients to obtain coverage without a prescription. 20 , 21 In the meantime, concerted efforts are needed to educate the public on their contraceptive options so that they can successfully obtain and continue using different methods as desired. These data suggest that expansion of contraceptive sources may provide significant inroads for the ease, convenience, and successful use of popular contraceptive methods. This study has limitations. We did not distinguish between preference for synchronous and asynchronous telehealth, which could result in improved measurement. While previous research indicates that preferred contraceptive source may differ across racial and ethnic groups, we lacked sufficient sample sizes to explore such differences. Our findings pertain only to individuals who currently used the pill, patch, and ring and do not address preferred source of contraception among the general population of individuals who could potentially use these methods, including those who may have faced the greatest access barriers and not been able to successfully obtain a desired SARC method, or individuals whose previous SARC use may have been supported by access to alternative sources. Finally, like all survey research, this study is limited in its inability to truly contextualize findings, even with the use of follow-up questions. Future studies investigating sources of contraception should examine how use and preferences change over time, especially as OTC access improves.

Advancing SRH equity requires person-centered strategies for contraceptive access. 31 The low level of preference for in-person care suggests that expanding contraceptive sources outside of traditional health care settings has a role in ameliorating barriers to access and can promote reproductive autonomy. Policymakers can also ensure equitable access to these methods by requiring all payers to reimburse clinicians in alternative settings, expand reimbursement for more types of clinicians, implement 1-year dispensing, and ensure that OTC methods are fully covered without a prescription.

Despite advances in method availability and insurance coverage, barriers to contraceptive access persist in the US, including logistical and interpersonal challenges, 1 misinformation about contraception, 2 , 3 , 4 and legal obstacles. 5 , 6 , 7 Individuals who wish to continue use of short-acting, reversible contraception (SARC, including the pill, patch, and ring) must periodically engage with the health care system to renew prescriptions and obtain additional supplies. Given widespread use of SARC 8 and requisite actions to continue use, access barriers have implications for reproductive autonomy. How and where people receive contraception—including location, frequency, and clinician type—influence whether people receive care that meets their needs. 9 , 10 , 11 In population-based data from Arizona, New Jersey, and Wisconsin, one-fourth of respondents expressed interest solely in traditional clinic-based contraceptive care, while most were interested in pharmacy-based and telehealth options. 12 In another study, 40% of females who used contraception were not obtaining care through their preferred approach. 13 Options for obtaining SARC have expanded beyond the traditional in-person, facility-based approach. Increasing evidence supports the safety and acceptability of alternative approaches, such as telehealth, pharmacist prescribing, and over-the-counter (OTC) provision. 14 During the COVID-19 pandemic, telehealth (patient-clinician visits conducted virtually) for contraception became more common. 15 , 16 Exclusively online, direct-to-consumer services for contraception have also proliferated. 17 , 18 As of August 2023, 28 states and Washington DC authorized pharmacists to prescribe hormonal contraception. 19 Additionally, an OTC progestin-only oral contraceptive pill became available in March 2024. 20 , 21 Advancing health equity demands expanding access points for contraception. Alternative sources of care that allow individuals to obtain contraception without visiting a clinic can support improved access for people residing in remote areas, whose schedules limit attendance of in-person visits, and who prefer to avoid interactions with the health care system due to racism, other historic abuses, and prior low quality of care. 22 , 23 , 24 , 25 Past and ongoing abuses include those specific to contraception (eg, forced and coerced sterilization, nonconsensual testing of contraceptives) 26 , 27 , 28 and more general (eg, denials of care based on race, algorithmic biases, and poorer quality of interpersonal care). 29 In a 2022 national survey, 67% of individuals assigned female sex at birth preferred to obtain contraception from a doctor’s office, but this proportion was substantially lower among respondents who were Black (19%) and Hispanic (18%), had low incomes (17%), and were uninsured (21%). 13 Holistic, person-centered contraceptive access, including the source of care, is a crucial element of reproductive self-determination. 30 , 31 , 32 Current approaches to measuring contraceptive access are not person-centered, focusing on use and services provided. 33 , 34 , 35 , 36 In this study, we aimed to describe the use of preferred source of contraception among current SARC users as a person-centered metric of contraceptive access. Additionally, we examined associations between previous experiences obtaining contraception and preference for traditional vs alternative contraceptive sources, hypothesizing that those who had experienced access barriers or low-quality care would be more likely to prefer alternative sources.